Skip to content
Tutorials Center
My learning

Topics

Lesson 4 of 38 ยท Getting Started

Degrees of Freedom and Workspace

Understand degrees of freedom, revolute and prismatic joints, the Grubler formula, configuration space and reachable workspace, with Python plots of a two-link arm.

How many ways can a robot move? That is its degrees of freedom (DOF): the number of independent quantities needed to say exactly where every part of it is. A door on hinges has one DOF (its angle). A free-floating drone has six. DOF decides what a robot can reach, how hard it is to control and how many motors it needs, so this is the first quantity to compute for any design.

Joints and DOF

Rigid objects in 3D space have 6 DOF: three positions (x, y, z) and three rotations (roll, pitch, yaw). In 2D a body has 3: x, y and heading. A joint removes freedom by constraining how two links move relative to each other.

Joint Motion allowed DOF
Revolute (hinge) rotation about one axis 1
Prismatic (slider) translation along one axis 1
Cylindrical rotate and slide on the same axis 2
Universal rotation about two axes 2
Spherical (ball) rotation about three axes 3

For a serial arm (a chain with one joint after another), the arm's DOF is simply the number of 1-DOF joints. A six-joint articulated arm has 6 DOF, which is exactly enough to place its tool at any position and orientation inside its reach. An arm with fewer joints cannot do it; one with more is redundant, which gives the choice of several postures for the same tool pose.

Counting DOF for general mechanisms

Mechanisms with loops (parallel robots, four-bar linkages) need a formula. The Grubler-Kutzbach formula for a planar mechanism with N links (including the fixed ground), J one-DOF joints:

DOF = 3 (N - 1 - J) + J

For spatial mechanisms the constant 3 becomes 6:

DOF = 6 (N - 1 - J) + sum of joint freedoms

Python Counting DOF with the Grubler formula

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The four-bar has 1 DOF: move one link and the others follow. The Stewart platform comes out at 6 even though it has 18 joints, because loops cancel freedoms.

Task space and joint space

Two descriptions of the robot's state matter:

  • Joint space (or configuration space): the vector of joint values, for example (ฮธ1, ฮธ2) for a two-link arm.
  • Task space (or operational space): where the end effector is, for example (x, y) in a plane or a full 6-D pose.

Forward kinematics maps joint space to task space; inverse kinematics goes the other way. The two spaces can have different sizes. A planar arm with three joints has a 3-D joint space but needs only 2 numbers (x, y) to place its tip, so the extra DOF can be used to avoid an obstacle.

Workspace

The workspace is the set of points the end effector can reach. The reachable workspace is every point it can touch in any orientation; the smaller dexterous workspace is where it can reach with every orientation. Joint limits, link lengths and collisions shape it. A planar two-link arm with link lengths l1 and l2 reaches an annulus (ring) between radius |l1 - l2| and l1 + l2.

Python Plotting the workspace of a 2-link arm

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The elbow limit removes the inner part of the ring; real arms have such limits because links collide or cables twist.

Singularities and redundancy

Even inside the workspace, some configurations are singular: the arm loses the ability to move its tip in some direction, for instance when it is fully stretched out. Motors may then need enormous speeds to produce small tip motion. The Jacobian lesson shows how to detect this. A redundant robot can often steer around a singularity using its extra joints.

Mobile robots: holonomic or not

Constraints are not always about joints. A car-like robot has 3 DOF of pose (x, y, heading) but only 2 controls (speed and steering), and it cannot slide sideways. Such constraints are non-holonomic: the robot can reach any pose, but not by any path (think parallel parking). Differential-drive robots and cars are non-holonomic; omnidirectional robots are holonomic. See Mobile Robot Kinematics.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. How many DOF does a rigid body moving freely in 3D space have?

    Three translations and three rotations.
  2. 2. A serial arm has 7 revolute joints. What is it called relative to a 6-DOF task?

    More joints than the task needs gives redundancy: several postures give the same tool pose.
  3. 3. What does the planar Grubler formula give for a four-bar linkage (4 links including ground, 4 revolute joints)?

    3 * (4 - 1 - 4) + 4 = 1.
  4. 4. What is a non-holonomic constraint?

    Non-holonomic robots can reach any pose but only along allowed paths.

Summary

  • DOF is the number of independent coordinates that describe a robot's configuration.
  • Each revolute or prismatic joint adds one DOF to a serial arm; loops need the Grubler formula.
  • Joint space and task space are different descriptions; kinematics converts between them.
  • The workspace is where the tool can reach; singularities and joint limits shape it.
  • Non-holonomic robots, such as cars, have fewer controls than pose DOF.

Next: Rotations and Transforms